Traceable Automatic Feature Transformation via Cascading Actor-Critic Agents

TL;DR

This paper introduces a novel traceable automatic feature transformation method using cascading actor-critic agents, improving feature space quality and model performance in high-dimensional data scenarios.

Contribution

It formulates feature transformation as an iterative process combining feature generation and selection, addressing limitations of existing methods with a new reinforcement learning approach.

Findings

24.7% improvement in F1 scores over state-of-the-art methods

Robustness demonstrated in high-dimensional data

Effective integration of local and global feature information

Abstract

Feature transformation for AI is an essential task to boost the effectiveness and interpretability of machine learning (ML). Feature transformation aims to transform original data to identify an optimal feature space that enhances the performances of a downstream ML model. Existing studies either combines preprocessing, feature selection, and generation skills to empirically transform data, or automate feature transformation by machine intelligence, such as reinforcement learning. However, existing studies suffer from: 1) high-dimensional non-discriminative feature space; 2) inability to represent complex situational states; 3) inefficiency in integrating local and global feature information. To fill the research gap, we formulate the feature transformation task as an iterative, nested process of feature generation and selection, where feature generation is to generate and add new features based on original features, and feature selection is to remove redundant features to control the size of feature space. Finally, we present extensive experiments and case studies to illustrate 24.7\% improvements in F1 scores compared with SOTAs and robustness in high-dimensional data.